A lattice quantum chromodynamics calculation has reduced the uncertainty in the strong coupling, a fundamental input for predictions involving quarks and gluons, to five parts per mille. The result uses low-energy information and can be applied to collider calculations without first being fitted to collider data, according to an account in CERN Courier.
The strong coupling, written as alpha-s, describes the strength of interactions between quarks and gluons. It enters calculations for many processes at proton colliders because protons are composite particles whose quarks are bound by gluons. Higgs production through gluon fusion, for example, depends strongly on the value: its probability is proportional to the square of alpha-s.
Precision has become increasingly important as measurements at the Large Hadron Collider reach the percentage level. An uncertain coupling can limit how closely theory predictions can be compared with experimental results. That problem will become more pronounced for the High-Luminosity LHC and proposed future facilities seeking small deviations that could point to physics beyond the Standard Model.
Unlike the electromagnetic coupling, alpha-s changes substantially with the distance or momentum scale being examined. Gluons carry the colour charge associated with the force and interact with one another. As a result, the coupling becomes weaker at short distances and high momentum transfers, a behaviour called asymptotic freedom, while it grows at lower energies.
Physicists conventionally quote alpha-s at the mass of the Z boson, 91.2 GeV. CERN Courier gives the Particle Data Group world average at that scale as 0.1180 plus or minus 0.0009, equivalent to a precision of 7.6 parts per mille. The new lattice result narrows the uncertainty to five parts per mille.
Lattice QCD is needed because ordinary perturbative expansions stop being useful when the interaction becomes strong at low energies. The method represents space and time as a finite four-dimensional grid, allowing computers to sample configurations of quark and gluon fields. Calculations must then control the effects of the finite grid spacing and volume before connecting the simulation to continuous physical space.
The approach has developed over decades. Early simulations were constrained by computing power and often omitted the influence of virtual quark loops, introducing an uncontrolled approximation. More realistic simulations with dynamical quarks became practical in the 2000s, and groups now compare and average lattice results through the Flavour Lattice Averaging Group.
The improved coupling does not change the theory of the strong interaction. Its significance is practical: a better-constrained input reduces one source of uncertainty in predictions, making increasingly precise collider measurements more useful as tests of the Standard Model.
